# Corpora Welcome to the FormosanBank Corpora section! Here, you’ll find comprehensive documentation of the corpora used in FormosanBank. Each corpus in this collection represents a unique linguistic dataset, encompassing various types of text and audio recordings. Our corpora are designed to support linguistic research, language education, and revitalization efforts, making these endangered languages accessible and analyzable for researchers, educators, and community members alike. Below is a list of the current corpora of which FormosanBank consists: * [ePark](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/epark) (text, audio, English, Mandarin) * [ILRDF Dictionaries](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/ilrdf-dictionaries) (text, audio, Mandarin) * [Wikipedias](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/wikipedias) (text) * [Presidential Apologies](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/presidential-apologies) (text, English, Mandarin) * [NTU Paiwan ASR](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/ntu-paiwan-asr) (text, audio) * [Virginia Fey's Amis Dictionary](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/virginia-feys-amis-dictionary) (text, English, Mandarin) * [Paiwan Stories](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/paiwan-stories) (text, audio, Mandarin) * [Rau-Dong](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/raudong) (text, glossing, English, Mandarin) * [The Montgomery Texts](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/montgomerytexts) (text, English) * [The Wakelin Texts](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/wakelintexts) (text, English, segmented) * [100 Paiwan Texts](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/hundredpaiwantexts) (text, glossing, English) * [FormosanBank GitBook](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/formosanbankgitbook) * [SEALS 33](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/seals33) (text, Mandarin, English) * [Glosbe](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/glosbe) (text, Mandarin (Traditional), Mandarin (Simplified)) * [Whitehorn Collection](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/whitehorn-collection) (audio) * [Yedda Palemeq Blog](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/yeddapalemeqblog) (text, English, segmentation, audio) * [Siraya Gospels](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/siraya-gospels) (text, English, Mandarin, Dutch) * [Wilang Yutas Videos](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/wilangyutasvideos) (text, Mandarin, audio) * [Tang Recordings of Taroko](https://ai4commsci.gitbook.io/formosanbank/the-bank-architecture/corpora/tangrecordingsoftaroko) (audio) * [NTU Corpus of Formosan Languages](https://github.com/FormosanBank/FormosanBankGitbook/blob/en-us/en-us/the-bank-architecture/corpora/NTUFormosanCorpus.md) (text, audio, glossing) *** ### Corpus Statistics #### By language | | Truku | Amis | Bunun | Kavalan | Rukai | Siraya | Paiwan | Puyuma | Thao | Saaroa | Sakizaya | Yami | Atayal | Seediq | Tsou | Kanakanabu | Saisiyat | | ------------------------- | ------ | --------- | ------- | ------- | ------- | ------ | ------- | ------- | ------- | ------ | --------- | ------- | ------- | ------- | ------ | ---------- | -------- | | Word count | 86,274 | 2,122,310 | 248,279 | 127,162 | 273,600 | 42,107 | 398,682 | 229,064 | 101,856 | 63,097 | 1,445,857 | 119,240 | 747,479 | 970,779 | 88,102 | 108,814 | 101,203 | | Total audio | 42.2h | 60.8h | 62.7h | 26.9h | 77.1h | 0 | 74.9h | 59.9h | 20.7h | 23.0h | 24.5h | 20.3h | 83.5h | 46.1h | 20.5h | 32.6h | 23.5h | | Transcribed | 20.8h | 60.6h | 62.7h | 26.9h | 77.1h | 0 | 62.4h | 59.9h | 20.7h | 23.0h | 24.5h | 20.3h | 77.3h | 46.0h | 20.5h | 32.6h | 23.5h | | Untranscribed | 21.4h | 0.2h | 0 | 0 | 0 | 0 | 12.5h | 0 | 0 | 0 | 0 | 0 | 6.2h | 0.1h | 0 | 0 | 0 | | Translated sentences | | | | | | | | | | | | | | | | | | | English | 2,187 | 13,870 | 13,759 | 4,544 | 15,175 | 1,951 | 12,357 | 8,687 | 2,155 | 2,183 | 3,105 | 3,062 | 13,575 | 8,135 | 3,132 | 4,598 | 3,497 | | Mandarin | 11,995 | 47,146 | 43,009 | 18,666 | 51,172 | 1,947 | 32,953 | 35,349 | 13,902 | 11,902 | 14,850 | 14,712 | 46,182 | 28,402 | 11,518 | 17,402 | 14,820 | | Morphologically segmented | 0 | 756 | 2,975 | 2,394 | 2,252 | 0 | 3,567 | 0 | 0 | 0 | 1,530 | 960 | 489 | 2,098 | 949 | 3,013 | 1,294 | | Proportion glossed | 0% | 100% | 100% | 100% | 100% | 0% | 82% | 0% | 0% | 0% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | #### By dialect | |

Amis
Coastal

|

Amis
Hengchun

|

Amis
Malan

|

Amis
Southern

|

Amis
UK

|

Amis
Xiuguluan

|

Bunun
Junqun

|

Bunun
Kaqun

|

Bunun
Luanqun

|

Bunun
Tanqun

|

Bunun
Zhuoqun

|

Rukai
Dawu

|

Rukai
Dona

|

Rukai
Eastern

|

Rukai
Maolin

|

Rukai
Wanshan

|

Rukai
Wutai

|

Paiwan
Central

|

Paiwan
Eastern

|

Paiwan
North Western

|

Paiwan
Northern

|

Paiwan
Sothern

|

Paiwan
Southern

|

Paiwan
UK

|

Puyuma
Jianhe

|

Puyuma
Nanwang

|

Puyuma
Xiqun

|

Puyuma
Zhiben

|

Atayal
FourSeasons

|

Atayal
Sekolik

|

Atayal
UK

|

Atayal
Wanda

|

Atayal
Wenshui

|

Atayal
YilanZeaol

|

Atayal
Zeaol

|

Seediq
DeluValley

|

Seediq
Duda

|

Seediq
Tegudaya

|

Seediq
UK

| | ------------------------- | ---------------------- | ----------------------- | -------------------- | ----------------------- | ----------------- | ------------------------ | ---------------------- | --------------------- | ----------------------- | ---------------------- | ----------------------- | -------------------- | -------------------- | ----------------------- | ---------------------- | ----------------------- | --------------------- | ------------------------ | ------------------------ | ------------------------------ | ------------------------- | ------------------------ | ------------------------- | ------------------- | ----------------------- | ------------------------ | ---------------------- | ----------------------- | ---------------------------- | ------------------------ | ------------------- | ---------------------- | ------------------------ | --------------------------- | ---------------------- | --------------------------- | --------------------- | ------------------------- | ------------------- | | Word count | 66,343 | 36,750 | 36,137 | 36,328 | 1,863,357 | 83,395 | 129,475 | 29,654 | 30,595 | 29,721 | 28,834 | 29,992 | 28,565 | 30,151 | 27,693 | 24,350 | 132,849 | 41,624 | 50,297 | 0 | 127,754 | 0 | 53,423 | 125,584 | 35,383 | 116,881 | 37,157 | 39,643 | 38,172 | 138,146 | 417,534 | 31,056 | 42,863 | 40,229 | 39,479 | 39,861 | 41,788 | 134,300 | 754,830 | | Total audio | 14.5h | 10.2h | 8.4h | 9.2h | 0.2h | 18.3h | 30.2h | 8.0h | 7.9h | 8.6h | 8.1h | 9.5h | 8.9h | 9.4h | 8.8h | 8.4h | 32.2h | 10.2h | 12.8h | 7.9h | 27.7h | 0.0h | 11.8h | 4.5h | 9.4h | 29.8h | 9.9h | 10.8h | 9.2h | 34.2h | 0.5h | 8.8h | 10.6h | 10.3h | 9.9h | 11.7h | 10.1h | 24.3h | 0.1h | | Transcribed | 14.5h | 10.2h | 8.4h | 9.2h | 0 | 18.3h | 30.2h | 8.0h | 7.9h | 8.6h | 8.1h | 9.5h | 8.9h | 9.4h | 8.8h | 8.4h | 32.2h | 10.2h | 12.8h | 0 | 27.7h | 0 | 11.7h | 0 | 9.4h | 29.8h | 9.9h | 10.8h | 9.2h | 28.5h | 0 | 8.8h | 10.6h | 10.3h | 9.9h | 11.7h | 10.1h | 24.3h | 0 | | Untranscribed | 0 | 0 | 0 | 0 | 0.2h | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 7.9h | 0 | 0.0h | 0.1h | 4.5h | 0 | 0 | 0 | 0 | 0 | 5.7h | 0.5h | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.1h | | Translated sentences | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | English | 2,910 | 2,155 | 2,144 | 2,158 | 40 | 4,463 | 5,132 | 2,154 | 2,163 | 2,154 | 2,156 | 2,151 | 2,154 | 2,149 | 2,158 | 2,150 | 4,413 | 2,721 | 2,400 | 0 | 3,197 | 0 | 3,968 | 71 | 2,153 | 2,150 | 2,229 | 2,155 | 2,159 | 2,309 | 0 | 2,154 | 2,644 | 2,160 | 2,149 | 2,159 | 2,158 | 3,802 | 16 | | Mandarin | 8,276 | 6,325 | 6,276 | 6,076 | 5,860 | 14,333 | 18,880 | 6,109 | 6,142 | 5,932 | 5,946 | 6,283 | 6,193 | 6,031 | 6,149 | 6,295 | 20,221 | 6,467 | 6,387 | 0 | 13,460 | 0 | 6,639 | 0 | 6,502 | 15,618 | 6,581 | 6,648 | 6,714 | 14,404 | 0 | 5,896 | 6,523 | 6,276 | 6,369 | 6,367 | 6,264 | 15,743 | 28 | | Morphologically segmented | 756 | 0 | 0 | 0 | 0 | 0 | 2,975 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2,252 | 537 | 157 | 0 | 995 | 0 | 1,807 | 71 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 489 | 0 | 0 | 0 | 0 | 2,098 | 0 | | Proportion glossed | 100% | 0% | 0% | 0% | 0% | 0% | 100% | 0% | 0% | 0% | 0% | 0% | 0% | 0% | 0% | 0% | 100% | 100% | 100% | 0% | 100% | 0% | 64% | 100% | 0% | 0% | 0% | 0% | 0% | 0% | 0% | 0% | 100% | 0% | 0% | 0% | 0% | 100% | 0% | *** ### Coming Soon In addition to the wide range of corpora already incorporated into FormosanBank, there is a large number of further corpora that permission to include in FormosanBank has been obtained and they are being processed at the moment. Below are some of these corpora: * NTU Corpus (Various languages) * Matthew's Gospel and John's Gospel (Siraya) * The Sedik Language of Formosa by Erin Asai (Seediq) * Chang's Seediq Reference Grammar (Seediq) * Chang's Kavalan reference grammar (Kavalan) * Poinsot Amis Dictionary (Amis) * Moedict Amis (Amis) * Asai's Seediq Language of Formosan (Seediq) * Wilang Yutas videos (Atayal) * ​hala saku la (videos - Atayal) * ​hala saku la (text - Atayal) * Tung's Descriptive study of Tsou (Tsou) * Jeng (1992) Topic and focus in Bunun (Bunun) * Blust's Thao Dictionary (Thao)